Why This Study Matters
John Henderson asks whether education itself causes people to participate more in politics — specifically, whether increases in schooling led to higher voting in the American South. The question speaks directly to a long-running debate in political behavior: education and turnout are strongly correlated, but can that link be interpreted causally?
A Natural Experiment in Public Health
Henderson exploits the Rockefeller Sanitary Commission’s (RSC) anti-hookworm campaign in the early twentieth-century American South as a plausibly exogenous shock to schooling. The RSC’s efforts to diagnose and treat hookworm infection reduced a disease that had depressed school attendance, thereby expanding primary and secondary education in some places more than others.
Two Measures of RSC Intervention
- Whether counties were exposed to the RSC campaign at all (binary exposure).
- The proportion of the local population actually treated by the campaign (a continuous measure).
This dual approach lets the author separate the presence of a public-health program from its intensity on the ground.
How the Analysis Works
- Henderson uses genetic matching to balance counties on observable factors that influenced where treatment happened, addressing the haphazard placement and uptake of RSC activities.
- He implements recently developed matching techniques tailored for continuous interventions to handle the proportion-treated measure.
- A suite of robustness checks and alternative specifications probe whether non-educational mechanisms or remaining imbalances could explain the results.
Key Findings
- Across specifications, the evidence points to a positive effect of increased schooling — driven by the hookworm campaign — on political participation.
- The results are robust to different matching procedures and to attempts to rule out alternative explanations tied to the RSC’s presence.
What This Means for Political Science
The paper provides new historical and causal evidence that expanding access to education can increase voter participation. By leveraging a public-health intervention as an exogenous source of schooling variation and by applying careful matching methods, Henderson strengthens the case that the well-known education–participation correlation reflects a substantive, causal education effect rather than simple selection or confounding.
